Hybrid simulation of a computing solution in a cloud computing environment with a simplified computing solution and a simulation model

Abstract

A simulation mechanism manages deployment of a simplified computing solution (SCS) and a corresponding simulation model that simulates a scaled multiple of the SCS to allow a deployment configuration for a large scale computing solution to be determined and tested before actually deploying the large scale computing solution.

The invention claimed is: 
     
       1. A computer-implemented method for simulation of a large scale computing solution to determine and test deployment of the large scale computing solution before full deployment, the method comprising:
 creating a computing solution for deployment on a physical system of computer resources with an initial state of a deployment configuration, where the computing solution is a special version of the large scale computing solution and has an element corresponding to each component of the large scale computing solution and each element includes at least one open virtualization format (OVF) file; 
 creating a simulation model corresponding to the computing solution wherein the simulation model includes a plurality of model elements, each of the plurality of model elements corresponding to one of each element of the computing solution that simulates a load on the system for the corresponding element; 
 running a simulation; 
 running a hybrid simulation by directly deploying both the computing solution and the simulation model together to the physical system of computer resources; 
 dynamically scaling up the computing solution with the simulation model by steps while running benchmark tests at each step, monitoring system performance of the physical system deploying the computing solution while running the benchmark tests and adjusting the deployment configuration; and 
 deploying the large scale computing solution with the adjusted deployment configuration determined by scaling up the simulation model. 
 
     
     
       2. The method of  claim 1  wherein the step of dynamically scaling up the computing solution with the simulation model further comprises:
 monitoring performance of the computing solution and adjusting the simulation model to make an adjusted simulation model; 
 deploying the simulation model; 
 running the benchmark tests on the adjusted simulation model; 
 where the benchmark tests are within performance parameters deploying the adjusted simulation model to the computing solution. 
 
     
     
       3. The method of  claim 1  wherein the large scale computing solution is deployed to a cloud computing environment. 
     
     
       4. The method of  claim 1  wherein the at least one OVF file comprises network resources, disk resources, memory resources, operating system resources, and computer processor resources allocated to components of the computing solution. 
     
     
       5. The method of  claim 4  wherein the computer processor resources include a number of physical computer processing units assigned to a corresponding component. 
     
     
       6. The method of  claim 1  wherein the step of dynamically scaling up the computing solution with the simulation model further comprises:
 determining a target scale, wherein the target scale represents the scale or ratio of the large scale computing solution to be deployed and the simulation model; 
 determining a step size; 
 deploying the computing solution; 
 monitoring the computing solution; 
 adjusting the simulation model by the step size; 
 deploying the simulation model; 
 running a benchmark on the simulation model; 
 where the benchmark is not within performance parameters then return to the step of adjusting the simulation model by the step size; 
 where the benchmark is within performance parameters deploy the step size to the computing solution; 
 where the target size is not achieved then return to the step of adjusting the simulation model by the step size. 
 
     
     
       7. A computer-implemented method for simulation of a large scale computing solution to determine and test deployment of the large scale computing solution before full deployment, the method comprising:
 creating a computing solution for deployment on a physical system of computer resources with an initial state of a deployment configuration, where the computing solution is a special version of the large scale computing solution and has an element corresponding to each component of the large scale computing solution and each element includes at least one open virtualization format (OVF) file; 
 creating a simulation model corresponding to the computing solution wherein the simulation model includes a plurality of model elements, each of the plurality of model elements corresponding to one of each element of the computing solution that simulates a load on the system for the corresponding element; 
 running a hybrid simulation by deploying both the computing solution and the simulation model together to the physical system of computer resources; 
 dynamically scaling up the computing solution with the simulation model by steps while running benchmark tests at each step, monitoring system performance of the system deploying the computing solution while running the benchmark tests and adjusting the deployment configuration; and 
 deploying the large scale computing solution with the adjusted deployment configuration determined by scaling up the simulation model; 
 wherein dynamically scaling up the simplified computing solution with the simulation model comprises:
 determining a target scale, wherein the target scale represents the scale or ratio of the large scale computing solution to be deployed and the simulation model; 
 determining a step size; 
 deploying the simplified computing solution to the physical system of computer resources; 
 monitoring the simplified computing solution and adjusting the simulation model by the step size; 
 deploying the simulation model; 
 running a benchmark on the simulation model; 
 where the benchmark is not within performance parameters then return to the step of adjusting the simulation model; 
 where the benchmark is within performance parameters deploy the step size to the computing solution; 
 where the target size is not achieved then return to the step of adjusting the simulation model by a step size. 
 
 
     
     
       8. The method of  claim 7  wherein the OVF file comprises network resources, disk resources, memory resources, operating system resources, and computer processor resources allocated to components of the computing solution. 
     
     
       9. The method of  claim 8  wherein the computer resources are in a cloud computing environment. 
     
     
       10. The method of  claim 8  wherein the computer processor resources include a number of physical computer processing units assigned to a corresponding component.